How to use Grok Bot and Claude Code to build a product-recommendation site
@emilylai reports building a product-recommendation site in seven hours with 94 prompts and 30 agents, using Grok Bot, Grok 4.7 and other agents for indexing, plus planning, creative/design and video roles. The post says it was built with Claude Code.
x.com
Suggested prompt
Suggested prompt written by Agents In Use, not the source author. Replace bracketed inputs and review tool permissions first.
Help me build a product-recommendation site with separate roles for indexing recommendations, planning, creative direction and design, video, and implementation with Claude Code.
Reach people exploring AI agents
See placements →Evidence at a glance
Sourced account- Source poster / publisher
- @emilylai
- Original account
- Read the source ↗
- Published
- Source link checked
- Reported result
- A product-recommendation site built in seven hours with 94 prompts and 30 agents, reported by the poster; independent build verification is not established.
- Human involvement
- @emilylai reports using 94 prompts. The post does not describe individual prompts, permissions, review gates or final approval steps.
A linked account supports attribution. It does not mean we independently reproduced its result.
Report an issue with this case
Open correction emailOpens your email app with the case link included. Nothing is sent automatically.
How the agent did it
- @emilylai used Grok Bot, Grok 4.7 and 19 other agents to index product recommendations, as stated in the post.
- The post assigns one agent to planning, seven to creative direction and design, and one to video.
- The site was built with Claude Code; the poster reports seven hours, 94 prompts and 30 agents overall.
@emilylai reports building lifechanging.emilylai.com over seven hours with 94 prompts and 30 agents. The post names Grok Bot and Grok 4.7 and says the site was built with Claude Code. It describes a division of jobs rather than a single agent handling every aspect of the project.
For indexing product recommendations, the wording is "Grok Bot, Grok 4.7 and 19 other agents". The post then lists one for planning, seven for creative direction and design, and one for video. Those are the roles as stated; this account does not recalculate the headcount, assign unnamed agents to products, or invent a more detailed coordination process.
The reported result is a site built in seven hours. The time, prompt count and agent count are reported by the poster, not independently measured here. The linked site is the project identified in the post, rather than evidence that every part of its implementation has been tested.
The account does not describe individual prompts, agent permissions, approval gates, costs or reliability checks. It supports a role-split website build with Claude Code, not a claim of hands-off delivery or ongoing autonomous operation.
Prompt
Suggested prompt (written by Agents In Use, not the author's)
Source: X · @emilylai
Frequently asked questions
How long did the site take to build?
Seven hours with 94 prompts and 30 agents, reported by the poster. These figures were not independently measured.
How does the post split the roles?
The post says Grok Bot, Grok 4.7 and 19 other agents indexed product recommendations, one handled planning, seven creative direction and design, and one video.
Which tool was used to build the site?
The post says it was built with Claude Code. It does not describe the individual coding prompts, deployment steps or approval gates.
More details
About this account
The post describes agents assigned to recommendation indexing, planning, creative direction and design, and video, with Claude Code used to build the site.
Read the full X post text, author handle, publication timestamp and profile bio via the FxTwitter mirror on 2026-10-26. Claims are reported by the poster, not independently reproduced. Prompt: Suggested, written by Agents In Use; the post does not share an exact task prompt.
Credit: @emilylai
Source published
Added to Agents In Use
Source checked
Tools and topics
Grok Bot · Grok 4.7 · Claude Code
Website building · Multi-agent workflow · Product recommendations
Editorial completeness: 100/100. Measures documented fields, not agent performance or verified outcomes.